DonorsChoose¶
DonorsChoose.org receives hundreds of thousands of project proposals each year for classroom projects in need of funding. Right now, a large number of volunteers is needed to manually screen each submission before it's approved to be posted on the DonorsChoose.org website.
Next year, DonorsChoose.org expects to receive close to 500,000 project proposals. As a result, there are three main problems they need to solve:
- How to scale current manual processes and resources to screen 500,000 projects so that they can be posted as quickly and as efficiently as possible
- How to increase the consistency of project vetting across different volunteers to improve the experience for teachers
- How to focus volunteer time on the applications that need the most assistance
The goal of the competition is to predict whether or not a DonorsChoose.org project proposal submitted by a teacher will be approved, using the text of project descriptions as well as additional metadata about the project, teacher, and school. DonorsChoose.org can then use this information to identify projects most likely to need further review before approval.
About the DonorsChoose Data Set¶
The train.csv data set provided by DonorsChoose contains the following features:
| Feature | Description |
|---|---|
project_id |
A unique identifier for the proposed project. Example: p036502 |
project_title |
Title of the project. Examples:
|
project_grade_category |
Grade level of students for which the project is targeted. One of the following enumerated values:
|
project_subject_categories |
One or more (comma-separated) subject categories for the project from the following enumerated list of values:
Examples:
|
school_state |
State where school is located (Two-letter U.S. postal code). Example: WY |
project_subject_subcategories |
One or more (comma-separated) subject subcategories for the project. Examples:
|
project_resource_summary |
An explanation of the resources needed for the project. Example:
|
project_essay_1 |
First application essay* |
project_essay_2 |
Second application essay* |
project_essay_3 |
Third application essay* |
project_essay_4 |
Fourth application essay* |
project_submitted_datetime |
Datetime when project application was submitted. Example: 2016-04-28 12:43:56.245 |
teacher_id |
A unique identifier for the teacher of the proposed project. Example: bdf8baa8fedef6bfeec7ae4ff1c15c56 |
teacher_prefix |
Teacher's title. One of the following enumerated values:
|
teacher_number_of_previously_posted_projects |
Number of project applications previously submitted by the same teacher. Example: 2 |
* See the section Notes on the Essay Data for more details about these features.
Additionally, the resources.csv data set provides more data about the resources required for each project. Each line in this file represents a resource required by a project:
| Feature | Description |
|---|---|
id |
A project_id value from the train.csv file. Example: p036502 |
description |
Desciption of the resource. Example: Tenor Saxophone Reeds, Box of 25 |
quantity |
Quantity of the resource required. Example: 3 |
price |
Price of the resource required. Example: 9.95 |
Note: Many projects require multiple resources. The id value corresponds to a project_id in train.csv, so you use it as a key to retrieve all resources needed for a project:
The data set contains the following label (the value you will attempt to predict):
| Label | Description |
|---|---|
project_is_approved |
A binary flag indicating whether DonorsChoose approved the project. A value of 0 indicates the project was not approved, and a value of 1 indicates the project was approved. |
Notes on the Essay Data¶
-
Prior to May 17, 2016, the prompts for the essays were as follows:
- __project_essay_1:__ "Introduce us to your classroom"
- __project_essay_2:__ "Tell us more about your students"
- __project_essay_3:__ "Describe how your students will use the materials you're requesting"
- __project_essay_4:__ "Close by sharing why your project will make a difference"
-
Starting on May 17, 2016, the number of essays was reduced from 4 to 2, and the prompts for the first 2 essays were changed to the following:
- __project_essay_1:__ "Describe your students: What makes your students special? Specific details about their background, your neighborhood, and your school are all helpful."
- __project_essay_2:__ "About your project: How will these materials make a difference in your students' learning and improve their school lives?"
For all projects with project_submitted_datetime of 2016-05-17 and later, the values of project_essay_3 and project_essay_4 will be NaN.
In [1]:
%matplotlib inline
import warnings
warnings.filterwarnings("ignore")
import sqlite3
import pandas as pd
import numpy as np
import nltk
import string
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.feature_extraction.text import TfidfTransformer
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.metrics import confusion_matrix
from sklearn import metrics
from sklearn.metrics import roc_curve, auc
from nltk.stem.porter import PorterStemmer
import re
# Tutorial about Python regular expressions: https://pymotw.com/2/re/
import string
from nltk.corpus import stopwords
from nltk.stem import PorterStemmer
from nltk.stem.wordnet import WordNetLemmatizer
from gensim.models import Word2Vec
from gensim.models import KeyedVectors
import pickle
from tqdm import tqdm
import os
from plotly import plotly
import plotly.offline as offline
import plotly.graph_objs as go
offline.init_notebook_mode()
from collections import Counter
1.1 Reading Data¶
In [3]:
project_data = pd.read_csv('train_data.csv')
resource_data = pd.read_csv('resources.csv')
#Loading top 4000 data points
#project_data = project_data.head(4000)
#resource_data = resource_data.head(4000)
In [4]:
print("Number of data points in train data", project_data.shape)
print('-'*50)
print("The attributes of data :", project_data.columns.values)
project_data.head()
Out[4]:
In [5]:
print("Number of data points in train data", resource_data.shape)
print(resource_data.columns.values)
resource_data.head(2)
Out[5]:
1.2 Data Analysis¶
In [6]:
# PROVIDE CITATIONS TO YOUR CODE IF YOU TAKE IT FROM ANOTHER WEBSITE.
# https://matplotlib.org/gallery/pie_and_polar_charts/pie_and_donut_labels.html#sphx-glr-gallery-pie-and-polar-charts-pie-and-donut-labels-py
y_value_counts = project_data['project_is_approved'].value_counts()
print("Number of projects thar are approved for funding ", y_value_counts[1], ", (", (y_value_counts[1]/(y_value_counts[1]+y_value_counts[0]))*100,"%)")
print("Number of projects thar are not approved for funding ", y_value_counts[0], ", (", (y_value_counts[0]/(y_value_counts[1]+y_value_counts[0]))*100,"%)")
fig, ax = plt.subplots(figsize=(6, 6), subplot_kw=dict(aspect="equal"))
recipe = ["Accepted", "Not Accepted"]
data = [y_value_counts[1], y_value_counts[0]]
wedges, texts = ax.pie(data, wedgeprops=dict(width=0.5), startangle=-40)
bbox_props = dict(boxstyle="square,pad=0.3", fc="w", ec="k", lw=0.72)
kw = dict(xycoords='data', textcoords='data', arrowprops=dict(arrowstyle="-"),
bbox=bbox_props, zorder=0, va="center")
for i, p in enumerate(wedges):
ang = (p.theta2 - p.theta1)/2. + p.theta1
y = np.sin(np.deg2rad(ang))
x = np.cos(np.deg2rad(ang))
horizontalalignment = {-1: "right", 1: "left"}[int(np.sign(x))]
connectionstyle = "angle,angleA=0,angleB={}".format(ang)
kw["arrowprops"].update({"connectionstyle": connectionstyle})
ax.annotate(recipe[i], xy=(x, y), xytext=(1.35*np.sign(x), 1.4*y),
horizontalalignment=horizontalalignment, **kw)
ax.set_title("Nmber of projects that are Accepted and not accepted")
plt.show()
1.2.1 Univariate Analysis: School State¶
In [7]:
# Pandas dataframe groupby count, mean: https://stackoverflow.com/a/19385591/4084039
temp = pd.DataFrame(project_data.groupby("school_state")["project_is_approved"].apply(np.mean)).reset_index()
# if you have data which contain only 0 and 1, then the mean = percentage (think about it)
temp.columns = ['state_code', 'num_proposals']
'''# How to plot US state heatmap: https://datascience.stackexchange.com/a/9620
scl = [[0.0, 'rgb(242,240,247)'],[0.2, 'rgb(218,218,235)'],[0.4, 'rgb(188,189,220)'],\
[0.6, 'rgb(158,154,200)'],[0.8, 'rgb(117,107,177)'],[1.0, 'rgb(84,39,143)']]
data = [ dict(
type='choropleth',
colorscale = scl,
autocolorscale = False,
locations = temp['state_code'],
z = temp['num_proposals'].astype(float),
locationmode = 'USA-states',
text = temp['state_code'],
marker = dict(line = dict (color = 'rgb(255,255,255)',width = 2)),
colorbar = dict(title = "% of pro")
) ]
layout = dict(
title = 'Project Proposals % of Acceptance Rate by US States',
geo = dict(
scope='usa',
projection=dict( type='albers usa' ),
showlakes = True,
lakecolor = 'rgb(255, 255, 255)',
),
)
fig = go.Figure(data=data, layout=layout)
offline.iplot(fig, filename='us-map-heat-map')
'''
Out[7]:
In [8]:
# https://www.csi.cuny.edu/sites/default/files/pdf/administration/ops/2letterstabbrev.pdf
temp.sort_values(by=['num_proposals'], inplace=True)
print("States with lowest % approvals")
print(temp.head(5))
print('='*50)
print("States with highest % approvals")
print(temp.tail(5))
print(temp.info())
In [9]:
#stacked bar plots matplotlib: https://matplotlib.org/gallery/lines_bars_and_markers/bar_stacked.html
def stack_plot(data, xtick, col2='project_is_approved', col3='total'):
ind = np.arange(data.shape[0])
plt.figure(figsize=(20,5))
p1 = plt.bar(ind, data[col3].values)
p2 = plt.bar(ind, data[col2].values)
plt.ylabel('Projects')
plt.title('Number of projects aproved vs rejected')
plt.xticks(ind, list(data[xtick].values))
plt.legend((p1[0], p2[0]), ('total', 'accepted'))
plt.show()
In [10]:
def univariate_barplots(data, col1, col2='project_is_approved', top=False):
# Count number of zeros in dataframe python: https://stackoverflow.com/a/51540521/4084039
temp = pd.DataFrame(project_data.groupby(col1)[col2].agg(lambda x: x.eq(1).sum())).reset_index()
# Pandas dataframe grouby count: https://stackoverflow.com/a/19385591/4084039
temp['total'] = pd.DataFrame(project_data.groupby(col1)[col2].agg({'total':'count'})).reset_index()['total']
temp['Avg'] = pd.DataFrame(project_data.groupby(col1)[col2].agg({'Avg':'mean'})).reset_index()['Avg']
temp.sort_values(by=['total'],inplace=True, ascending=False)
if top:
temp = temp[0:top]
stack_plot(temp, xtick=col1, col2=col2, col3='total')
print(temp.head(5))
print("="*50)
print(temp.tail(5))
In [11]:
univariate_barplots(project_data, 'school_state', 'project_is_approved', False)
SUMMARY: Every state has greater than 80% success rate in approval
1.2.2 Univariate Analysis: teacher_prefix¶
In [12]:
univariate_barplots(project_data, 'teacher_prefix', 'project_is_approved' , top=False)
SUMMARY: The teacher_prefix containing Dr. has least success rate and the remaining data contain around 80% success rate in approval
1.2.3 Univariate Analysis: project_grade_category¶
In [13]:
univariate_barplots(project_data, 'project_grade_category', 'project_is_approved', top=False)
Summary: From 109k data we considered, we can see that all the data existing under the column Project grade category has an average success rate greater than 83%.
1.2.4 Univariate Analysis: project_subject_categories¶
In [14]:
catogories = list(project_data['project_subject_categories'].values)
# remove special characters from list of strings python: https://stackoverflow.com/a/47301924/4084039
# https://www.geeksforgeeks.org/removing-stop-words-nltk-python/
# https://stackoverflow.com/questions/23669024/how-to-strip-a-specific-word-from-a-string
# https://stackoverflow.com/questions/8270092/remove-all-whitespace-in-a-string-in-python
cat_list = []
for i in catogories:
temp = ""
# consider we have text like this "Math & Science, Warmth, Care & Hunger"
for j in i.split(','): # it will split it in three parts ["Math & Science", "Warmth", "Care & Hunger"]
if 'The' in j.split(): # this will split each of the catogory based on space "Math & Science"=> "Math","&", "Science"
j=j.replace('The','') # if we have the words "The" we are going to replace it with ''(i.e removing 'The')
j = j.replace(' ','') # we are placeing all the ' '(space) with ''(empty) ex:"Math & Science"=>"Math&Science"
temp+=j.strip()+" " #" abc ".strip() will return "abc", remove the trailing spaces
temp = temp.replace('&','_') # we are replacing the & value into
cat_list.append(temp.strip())
In [15]:
project_data['clean_categories'] = cat_list
project_data.drop(['project_subject_categories'], axis=1, inplace=True)
project_data.head(2)
Out[15]:
In [16]:
univariate_barplots(project_data, 'clean_categories', 'project_is_approved', top=20)
SUMMARY: The project_subject_categories column has been cleaned with neccessary changes and renamed with cleancategories. From the above graph we see that the all combinations of categories have the approval rate around 80%.
In [17]:
# count of all the words in corpus python: https://stackoverflow.com/a/22898595/4084039
from collections import Counter
my_counter = Counter()
for word in project_data['clean_categories'].values:
my_counter.update(word.split())
In [18]:
# dict sort by value python: https://stackoverflow.com/a/613218/4084039
cat_dict = dict(my_counter)
sorted_cat_dict = dict(sorted(cat_dict.items(), key=lambda kv: kv[1]))
ind = np.arange(len(sorted_cat_dict))
plt.figure(figsize=(20,5))
p1 = plt.bar(ind, list(sorted_cat_dict.values()))
plt.ylabel('Projects')
plt.xlabel('Category')
plt.title('% of projects aproved category wise')
plt.xticks(ind, list(sorted_cat_dict.keys()))
plt.show()
In [19]:
for i, j in sorted_cat_dict.items():
print("{:20} :{:10}".format(i,j))
SUMMARY: From the data sample we considered, we can analyse that the projects related to Literacy_Language category have highest approval rate while the Warmth category have least approval rate.
1.2.5 Univariate Analysis: project_subject_subcategories¶
In [20]:
sub_catogories = list(project_data['project_subject_subcategories'].values)
# remove special characters from list of strings python: https://stackoverflow.com/a/47301924/4084039
# https://www.geeksforgeeks.org/removing-stop-words-nltk-python/
# https://stackoverflow.com/questions/23669024/how-to-strip-a-specific-word-from-a-string
# https://stackoverflow.com/questions/8270092/remove-all-whitespace-in-a-string-in-python
sub_cat_list = []
for i in sub_catogories:
temp = ""
# consider we have text like this "Math & Science, Warmth, Care & Hunger"
for j in i.split(','): # it will split it in three parts ["Math & Science", "Warmth", "Care & Hunger"]
if 'The' in j.split(): # this will split each of the catogory based on space "Math & Science"=> "Math","&", "Science"
j=j.replace('The','') # if we have the words "The" we are going to replace it with ''(i.e removing 'The')
j = j.replace(' ','') # we are placeing all the ' '(space) with ''(empty) ex:"Math & Science"=>"Math&Science"
temp +=j.strip()+" "#" abc ".strip() will return "abc", remove the trailing spaces
temp = temp.replace('&','_')
sub_cat_list.append(temp.strip())
In [21]:
project_data['clean_subcategories'] = sub_cat_list
project_data.drop(['project_subject_subcategories'], axis=1, inplace=True)
project_data.head(2)
Out[21]:
In [22]:
univariate_barplots(project_data, 'clean_subcategories', 'project_is_approved', top=50)
SUMMARY : When we consider subcategories, we can see that projects with all combinations of sub categories have an average approval rate greater than 80%.
In [23]:
# count of all the words in corpus python: https://stackoverflow.com/a/22898595/4084039
from collections import Counter
my_counter = Counter()
for word in project_data['clean_subcategories'].values:
my_counter.update(word.split())
In [24]:
# dict sort by value python: https://stackoverflow.com/a/613218/4084039
sub_cat_dict = dict(my_counter)
sorted_sub_cat_dict = dict(sorted(sub_cat_dict.items(), key=lambda kv: kv[1]))
ind = np.arange(len(sorted_sub_cat_dict))
plt.figure(figsize=(20,5))
p1 = plt.bar(ind, list(sorted_sub_cat_dict.values()))
plt.ylabel('Projects')
plt.xlabel('Sub Category')
plt.title('% of projects aproved subcategory wise')
plt.xticks(ind, list(sorted_sub_cat_dict.keys()))
plt.show()
In [25]:
for i, j in sorted_sub_cat_dict.items():
print("{:20} :{:10}".format(i,j))
1.2.6 Univariate Analysis: Text features (Title)¶
In [26]:
#How to calculate number of words in a string in DataFrame: https://stackoverflow.com/a/37483537/4084039
word_count = project_data['project_title'].str.split().apply(len).value_counts()
word_dict = dict(word_count)
word_dict = dict(sorted(word_dict.items(), key=lambda kv: kv[1]))
ind = np.arange(len(word_dict))
plt.figure(figsize=(20,5))
p1 = plt.bar(ind, list(word_dict.values()))
plt.ylabel('Numeber of projects')
plt.xlabel('Numeber words in project title')
plt.title('Words for each title of the project')
plt.xticks(ind, list(word_dict.keys()))
plt.show()
SUMMARY : From the above graph we can see that most of the submitted project titles has a word count of >=3 or we may also say the average word count is around 3-7.
In [27]:
approved_title_word_count = project_data[project_data['project_is_approved']==1]['project_title'].str.split().apply(len)
approved_title_word_count = approved_title_word_count.values
rejected_title_word_count = project_data[project_data['project_is_approved']==0]['project_title'].str.split().apply(len)
rejected_title_word_count = rejected_title_word_count.values
In [28]:
# https://glowingpython.blogspot.com/2012/09/boxplot-with-matplotlib.html
plt.boxplot([approved_title_word_count, rejected_title_word_count])
plt.xticks([1,2],('Approved Projects','Rejected Projects'))
plt.ylabel('Words in project title')
plt.grid()
plt.show()
In [29]:
plt.figure(figsize=(10,3))
sns.kdeplot(approved_title_word_count,label="Approved Projects", bw=0.6)
sns.kdeplot(rejected_title_word_count,label="Not Approved Projects", bw=0.6)
plt.legend()
plt.show()
SUMMARY : This above KDE plot give us a detailed view stating that the average word count in the project title is around 4 for both approved and not approved projects.
1.2.7 Univariate Analysis: Text features (Project Essay's)¶
In [30]:
# merge two column text dataframe:
project_data["essay"] = project_data["project_essay_1"].map(str) +\
project_data["project_essay_2"].map(str) + \
project_data["project_essay_3"].map(str) + \
project_data["project_essay_4"].map(str)
In [31]:
approved_word_count = project_data[project_data['project_is_approved']==1]['essay'].str.split().apply(len)
approved_word_count = approved_word_count.values
rejected_word_count = project_data[project_data['project_is_approved']==0]['essay'].str.split().apply(len)
rejected_word_count = rejected_word_count.values
In [32]:
# https://glowingpython.blogspot.com/2012/09/boxplot-with-matplotlib.html
plt.boxplot([approved_word_count, rejected_word_count])
plt.title('Words for each essay of the project')
plt.xticks([1,2],('Approved Projects','Rejected Projects'))
plt.ylabel('Words in project essays')
plt.grid()
plt.show()
In [33]:
plt.figure(figsize=(10,3))
sns.distplot(approved_word_count, hist=False, label="Approved Projects")
sns.distplot(rejected_word_count, hist=False, label="Not Approved Projects")
plt.title('Words for each essay of the project')
plt.xlabel('Number of words in each eassay')
plt.legend()
plt.show()
SUMMARY : The above two PDF's of the approved and not approved projects has an average word count around 200 in Project essay. Also a nice observation is that the the after the count 250 the project not approved line has been decline with the increase in the word frequency.
1.2.8 Univariate Analysis: Cost per project¶
In [34]:
# we get the cost of the project using resource.csv file
resource_data.head(2)
Out[34]:
In [35]:
# https://stackoverflow.com/questions/22407798/how-to-reset-a-dataframes-indexes-for-all-groups-in-one-step
price_data = resource_data.groupby('id').agg({'price':'sum', 'quantity':'sum'}).reset_index()
price_data.head(2)
Out[35]:
In [36]:
# join two dataframes in python:
project_data = pd.merge(project_data, price_data, on='id', how='left')
In [37]:
approved_price = project_data[project_data['project_is_approved']==1]['price'].values
rejected_price = project_data[project_data['project_is_approved']==0]['price'].values
In [38]:
# https://glowingpython.blogspot.com/2012/09/boxplot-with-matplotlib.html
plt.boxplot([approved_price, rejected_price])
plt.title('Box Plots of Cost per approved and not approved Projects')
plt.xticks([1,2],('Approved Projects','Rejected Projects'))
plt.ylabel('Price')
plt.grid()
plt.show()
In [39]:
# http://zetcode.com/python/prettytable/
from prettytable import PrettyTable
#If you get a ModuleNotFoundError error , install prettytable using: pip3 install prettytable
x = PrettyTable()
x.field_names = ["Percentile", "Approved Projects", "Not Approved Projects"]
for i in range(0,101,5):
x.add_row([i,np.round(np.percentile(approved_price,i), 3), np.round(np.percentile(rejected_price,i), 3)])
print(x)
1.2.9 Univariate Analysis: teacher_number_of_previously_posted_projects
Please do this on your own based on the data analysis that was done in the above cells
In [40]:
abc = project_data['teacher_number_of_previously_posted_projects']
In [41]:
abc.describe()
Out[41]:
In [42]:
approved_tppp = project_data[project_data['project_is_approved']==1]['teacher_number_of_previously_posted_projects'].values
rejected_tppp = project_data[project_data['project_is_approved']==0]['teacher_number_of_previously_posted_projects'].values
In [43]:
plt.figure(figsize=(10,3))
sns.distplot(approved_tppp, hist=False, label="Approved Projects")
sns.distplot(rejected_tppp, hist=False, label="Not Approved Projects")
plt.title('teachers previous projects posted approved and not approved Projects')
plt.xlabel('teachers previous projects posted')
plt.legend()
plt.show()
SUMMARY : Considering the above plot teachers having less number of posted projects has a very low chance of approval._
In [44]:
# https://glowingpython.blogspot.com/2012/09/boxplot-with-matplotlib.html
plt.boxplot([approved_tppp, rejected_tppp])
plt.title('Box Plots of teachers no of previously approved and not approved Projects')
plt.xticks([1,2],('Approved Projects','Rejected Projects'))
plt.ylabel('Teachers no of prev projects')
plt.grid()
plt.show()
In [45]:
univariate_barplots(project_data, 'teacher_number_of_previously_posted_projects', 'project_is_approved', False,)
1.2.10 Univariate Analysis: project_resource_summary
Please do this on your own based on the data analysis that was done in the above cells
Check if the presence of the numerical digits in the project_resource_summary effects the acceptance of the project or not. If you observe that presence of the numerical digits is helpful in the classification, please include it for further process or you can ignore it.
In [46]:
import re
In [47]:
project_data['project_res_sum_digits'] = 2
In [48]:
project_data['project_res_sum_digits'].tail()
Out[48]:
In [49]:
'''
from tqdm import tqdm
count = []
c1 = []
for line in tqdm(range(0,len(project_data['project_resource_summary']))):
if re.findall('[0-9]',(project_data['project_resource_summary'][line])):
project_data['project_res_sum_digits'][line] = 1
count.append(line)
else:
project_data['project_res_sum_digits'][line] = 0
c1.append(line)
'''
Out[49]:
In [50]:
project_data['project_res_sum_digits'] = list(map(lambda x : 1 if re.findall('[0-9]',x) else 0,tqdm(project_data['project_resource_summary'])))
In [51]:
from collections import Counter
Counter(project_data['project_res_sum_digits']).keys() # equals to list(set(words))
Counter(project_data['project_res_sum_digits']).values()
Out[51]:
In [52]:
project_data['project_res_sum_digits'].value_counts()
Out[52]:
In [53]:
project_data[project_data['project_res_sum_digits']==0]['project_resource_summary'][1]
Out[53]:
In [54]:
project_data['project_res_sum_digits'].unique()
Out[54]:
In [55]:
univariate_barplots(project_data, 'project_res_sum_digits', 'project_is_approved', False,)
SUMMARY : In the above count plot, we can see that the projects having digits in summary are in less count but such projects have more approval rate.
In [56]:
approved_prs = project_data[project_data['project_is_approved']==1]['project_res_sum_digits'].values
rejected_prs = project_data[project_data['project_is_approved']==0]['project_res_sum_digits'].values
In [57]:
project_data.groupby(['project_is_approved']).count()
Out[57]:
In [58]:
project_data[project_data['project_is_approved']==1]['project_res_sum_digits'].count()
Out[58]:
In [59]:
len(approved_prs)
Out[59]:
In [60]:
plt.figure(figsize=(10,3))
sns.distplot(approved_prs, hist=False, label="Approved Projects")
sns.distplot(rejected_prs, hist=False, label="Not Approved Projects")
plt.title('Project summary containing digits which are approved and not approved Projects')
plt.xlabel('Project summary containing digits which are posted of a project')
plt.legend()
plt.show()
In [61]:
sns.countplot(x='project_res_sum_digits',hue='project_is_approved',data=project_data)
Out[61]:
SUMMARY : Here using the count plot we distingushed about the approval and the not approved projects which contains a digit in project summary.
In [62]:
# http://zetcode.com/python/prettytable/
from prettytable import PrettyTable
#If you get a ModuleNotFoundError error , install prettytable using: pip3 install prettytable
x = PrettyTable()
x.field_names = ["Percentile", "Approved Projects", "Not Approved Projects"]
for i in range(0,101,5):
x.add_row([i,np.round(np.percentile(approved_prs,i), 3), np.round(np.percentile(rejected_prs,i), 3)])
print(x)
1.3 Text preprocessing¶
1.3.1 Essay Text¶
In [63]:
project_data.head(2)
Out[63]:
In [64]:
# printing some random essays.
print(project_data['essay'].values[0])
print("="*50)
print(project_data['essay'].values[150])
print("="*50)
print(project_data['essay'].values[1000])
print("="*50)
print(project_data['essay'].values[2000])
print("="*50)
print(project_data['essay'].values[3999])
print("="*50)
In [65]:
# https://stackoverflow.com/a/47091490/4084039
import re
def decontracted(phrase):
# specific
phrase = re.sub(r"won't", "will not", phrase)
phrase = re.sub(r"can\'t", "can not", phrase)
# general
phrase = re.sub(r"n\'t", " not", phrase)
phrase = re.sub(r"\'re", " are", phrase)
phrase = re.sub(r"\'s", " is", phrase)
phrase = re.sub(r"\'d", " would", phrase)
phrase = re.sub(r"\'ll", " will", phrase)
phrase = re.sub(r"\'t", " not", phrase)
phrase = re.sub(r"\'ve", " have", phrase)
phrase = re.sub(r"\'m", " am", phrase)
return phrase
In [66]:
sent = decontracted(project_data['essay'].values[2000])
print(sent)
print("="*50)
In [67]:
# \r \n \t remove from string python: http://texthandler.com/info/remove-line-breaks-python/
sent = sent.replace('\\r', ' ')
sent = sent.replace('\\"', ' ')
sent = sent.replace('\\n', ' ')
print(sent)
In [68]:
#remove spacial character: https://stackoverflow.com/a/5843547/4084039
sent = re.sub('[^A-Za-z0-9]+', ' ', sent)
print(sent)
In [69]:
# https://gist.github.com/sebleier/554280
# we are removing the words from the stop words list: 'no', 'nor', 'not'
stopwords= ['i', 'me', 'my', 'myself', 'we', 'our', 'ours', 'ourselves', 'you', "you're", "you've",\
"you'll", "you'd", 'your', 'yours', 'yourself', 'yourselves', 'he', 'him', 'his', 'himself', \
'she', "she's", 'her', 'hers', 'herself', 'it', "it's", 'its', 'itself', 'they', 'them', 'their',\
'theirs', 'themselves', 'what', 'which', 'who', 'whom', 'this', 'that', "that'll", 'these', 'those', \
'am', 'is', 'are', 'was', 'were', 'be', 'been', 'being', 'have', 'has', 'had', 'having', 'do', 'does', \
'did', 'doing', 'a', 'an', 'the', 'and', 'but', 'if', 'or', 'because', 'as', 'until', 'while', 'of', \
'at', 'by', 'for', 'with', 'about', 'against', 'between', 'into', 'through', 'during', 'before', 'after',\
'above', 'below', 'to', 'from', 'up', 'down', 'in', 'out', 'on', 'off', 'over', 'under', 'again', 'further',\
'then', 'once', 'here', 'there', 'when', 'where', 'why', 'how', 'all', 'any', 'both', 'each', 'few', 'more',\
'most', 'other', 'some', 'such', 'only', 'own', 'same', 'so', 'than', 'too', 'very', \
's', 't', 'can', 'will', 'just', 'don', "don't", 'should', "should've", 'now', 'd', 'll', 'm', 'o', 're', \
've', 'y', 'ain', 'aren', "aren't", 'couldn', "couldn't", 'didn', "didn't", 'doesn', "doesn't", 'hadn',\
"hadn't", 'hasn', "hasn't", 'haven', "haven't", 'isn', "isn't", 'ma', 'mightn', "mightn't", 'mustn',\
"mustn't", 'needn', "needn't", 'shan', "shan't", 'shouldn', "shouldn't", 'wasn', "wasn't", 'weren', "weren't", \
'won', "won't", 'wouldn', "wouldn't"]
In [70]:
# Combining all the above statemennts
from tqdm import tqdm
preprocessed_essays = []
# tqdm is for printing the status bar
for sentance in tqdm(project_data['essay'].values):
sent = decontracted(sentance)
sent = sent.replace('\\r', ' ')
sent = sent.replace('\\"', ' ')
sent = sent.replace('\\n', ' ')
sent = re.sub('[^A-Za-z0-9]+', ' ', sent)
# https://gist.github.com/sebleier/554280
sent = ' '.join(e for e in sent.split() if e not in stopwords)
preprocessed_essays.append(sent.lower().strip())
In [71]:
# after preprocesing
preprocessed_essays[2000]
Out[71]:
1.3.2 Project title Text
In [72]:
# similarly you can preprocess the titles also
In [73]:
pt = project_data['project_title']
In [74]:
pt.head()
Out[74]:
In [75]:
pt.nunique()
Out[75]:
In [76]:
pt.values[100]
Out[76]:
In [77]:
# https://stackoverflow.com/a/47091490/4084039
import re
def decontracted(phrase):
# specific
phrase = re.sub(r"won't", "will not", phrase)
phrase = re.sub(r"can\'t", "can not", phrase)
# general
phrase = re.sub(r"n\'t", " not", phrase)
phrase = re.sub(r"\'re", " are", phrase)
phrase = re.sub(r"\'s", " is", phrase)
phrase = re.sub(r"\'d", " would", phrase)
phrase = re.sub(r"\'ll", " will", phrase)
phrase = re.sub(r"\'t", " not", phrase)
phrase = re.sub(r"\'ve", " have", phrase)
phrase = re.sub(r"\'m", " am", phrase)
return phrase
In [78]:
sent = decontracted(pt.values[2000])
print(sent)
print("="*50)
In [79]:
# \r \n \t remove from string python: http://texthandler.com/info/remove-line-breaks-python/
sent = sent.replace('\\r', ' ')
sent = sent.replace('\\"', ' ')
sent = sent.replace('\\n', ' ')
print(sent)
In [80]:
#remove spacial character: https://stackoverflow.com/a/5843547/4084039
sent = re.sub('[^A-Za-z0-9]+', ' ', sent)
print(sent)
In [81]:
# https://gist.github.com/sebleier/554280
# we are removing the words from the stop words list: 'no', 'nor', 'not'
stopwords= ['i', 'me', 'my', 'myself', 'we', 'our', 'ours', 'ourselves', 'you', "you're", "you've",\
"you'll", "you'd", 'your', 'yours', 'yourself', 'yourselves', 'he', 'him', 'his', 'himself', \
'she', "she's", 'her', 'hers', 'herself', 'it', "it's", 'its', 'itself', 'they', 'them', 'their',\
'theirs', 'themselves', 'what', 'which', 'who', 'whom', 'this', 'that', "that'll", 'these', 'those', \
'am', 'is', 'are', 'was', 'were', 'be', 'been', 'being', 'have', 'has', 'had', 'having', 'do', 'does', \
'did', 'doing', 'a', 'an', 'the', 'and', 'but', 'if', 'or', 'because', 'as', 'until', 'while', 'of', \
'at', 'by', 'for', 'with', 'about', 'against', 'between', 'into', 'through', 'during', 'before', 'after',\
'above', 'below', 'to', 'from', 'up', 'down', 'in', 'out', 'on', 'off', 'over', 'under', 'again', 'further',\
'then', 'once', 'here', 'there', 'when', 'where', 'why', 'how', 'all', 'any', 'both', 'each', 'few', 'more',\
'most', 'other', 'some', 'such', 'only', 'own', 'same', 'so', 'than', 'too', 'very', \
's', 't', 'can', 'will', 'just', 'don', "don't", 'should', "should've", 'now', 'd', 'll', 'm', 'o', 're', \
've', 'y', 'ain', 'aren', "aren't", 'couldn', "couldn't", 'didn', "didn't", 'doesn', "doesn't", 'hadn',\
"hadn't", 'hasn', "hasn't", 'haven', "haven't", 'isn', "isn't", 'ma', 'mightn', "mightn't", 'mustn',\
"mustn't", 'needn', "needn't", 'shan', "shan't", 'shouldn', "shouldn't", 'wasn', "wasn't", 'weren', "weren't", \
'won', "won't", 'wouldn', "wouldn't"]
In [82]:
# Combining all the above statemennts
from tqdm import tqdm
preprocessed_titles = []
# tqdm is for printing the status bar
for sentance in tqdm(pt.values):
sent = decontracted(sentance)
sent = sent.replace('\\r', ' ')
sent = sent.replace('\\"', ' ')
sent = sent.replace('\\n', ' ')
sent = re.sub('[^A-Za-z0-9]+', ' ', sent)
# https://gist.github.com/sebleier/554280
sent = ' '.join(e for e in sent.split() if e not in stopwords)
preprocessed_titles.append(sent.lower().strip())
In [83]:
preprocessed_titles[2000:2010]
Out[83]:
1. 4 Preparing data for models¶
In [84]:
project_data.columns
Out[84]:
we are going to consider
- school_state : categorical data
- clean_categories : categorical data
- clean_subcategories : categorical data
- project_grade_category : categorical data
- teacher_prefix : categorical data
- project_title : text data
- text : text data
- project_resource_summary: text data
- quantity : numerical
- teacher_number_of_previously_posted_projects : numerical
- price : numerical1.4.1 Vectorizing Categorical data¶
In [85]:
# we use count vectorizer to convert the values into one hot encoded features
from sklearn.feature_extraction.text import CountVectorizer
vectorizer = CountVectorizer(vocabulary=list(sorted_cat_dict.keys()), lowercase=False, binary=True)
vectorizer.fit(project_data['clean_categories'].values)
print(vectorizer.get_feature_names())
categories_one_hot = vectorizer.transform(project_data['clean_categories'].values)
print("Shape of matrix after one hot encodig ",categories_one_hot.shape)
In [86]:
# we use count vectorizer to convert the values into one hot encoded features
vectorizer = CountVectorizer(vocabulary=list(sorted_sub_cat_dict.keys()), lowercase=False, binary=True)
vectorizer.fit(project_data['clean_subcategories'].values)
print(vectorizer.get_feature_names())
sub_categories_one_hot = vectorizer.transform(project_data['clean_subcategories'].values)
print("Shape of matrix after one hot encodig ",sub_categories_one_hot.shape)
In [87]:
# Please do the similar feature encoding with state, teacher_prefix and project_grade_category also
school state¶
In [88]:
# count of all the words in corpus python: https://stackoverflow.com/a/22898595/4084039
from collections import Counter
my_counter = Counter()
for word in project_data['school_state'].values:
my_counter.update(word.split())
In [89]:
# dict sort by value python: https://stackoverflow.com/a/613218/4084039
cat_dict = dict(my_counter)
sorted_scl_dict = dict(sorted(cat_dict.items(), key=lambda kv: kv[1]))
ind = np.arange(len(sorted_scl_dict))
plt.figure(figsize=(20,5))
p1 = plt.bar(ind, list(sorted_scl_dict.values()))
plt.ylabel('Projects')
plt.title('% of projects aproved school state wise')
plt.xticks(ind, list(sorted_scl_dict.keys()))
plt.show()
In [90]:
# we use count vectorizer to convert the values into one hot encoded features
vectorizer = CountVectorizer(vocabulary=list(sorted_scl_dict.keys()), lowercase=False, binary=True)
vectorizer.fit(project_data['school_state'].values)
print(vectorizer.get_feature_names())
sub_categories_one_hot_1 = vectorizer.transform(project_data['school_state'].values)
print("Shape of matrix after one hot encodig ",sub_categories_one_hot_1.shape)
teacher prefix¶
In [91]:
project_data.groupby(['teacher_prefix'])['teacher_prefix'].count()
Out[91]:
In [92]:
project_data['teacher_prefix'][project_data['teacher_prefix'].isnull()==True]
Out[92]:
In [93]:
project_data['teacher_prefix'].fillna(project_data['teacher_prefix'].mode()[0],inplace=True)
In [94]:
project_data['teacher_prefix'][project_data['teacher_prefix'].isnull()==True]
Out[94]:
In [95]:
# count of all the words in corpus python: https://stackoverflow.com/a/22898595/4084039
from collections import Counter
my_counter = Counter()
for word in project_data['teacher_prefix'].values:
my_counter.update(word.split())
In [96]:
# dict sort by value python: https://stackoverflow.com/a/613218/4084039
cat_dict = dict(my_counter)
sorted_tp_dict = dict(sorted(cat_dict.items(), key=lambda kv: kv[1]))
ind = np.arange(len(sorted_tp_dict))
plt.figure(figsize=(20,5))
p1 = plt.bar(ind, list(sorted_tp_dict.values()))
plt.ylabel('Projects')
plt.title('% of projects aproved teacher_prefix wise')
plt.xticks(ind, list(sorted_tp_dict.keys()))
plt.show()
In [97]:
# we use count vectorizer to convert the values into one hot encoded features
vectorizer = CountVectorizer(vocabulary=list(sorted_tp_dict.keys()), lowercase=False, binary=True)
vectorizer.fit(project_data['teacher_prefix'].values)
print(vectorizer.get_feature_names())
sub_categories_one_hot_2 = vectorizer.transform(project_data['teacher_prefix'].values)
print("Shape of matrix after one hot encodig ",sub_categories_one_hot_2.shape)
project grade category¶
In [98]:
project_data['project_grade_category'][project_data['project_grade_category'].isnull()==True]
Out[98]:
In [99]:
# count of all the words in corpus python: https://stackoverflow.com/a/22898595/4084039
from collections import Counter
my_counter = Counter()
for word in project_data['project_grade_category'].values:
my_counter.update(word.split())
In [100]:
# dict sort by value python: https://stackoverflow.com/a/613218/4084039
cat_dict = dict(my_counter)
sorted_tp_dict = dict(sorted(cat_dict.items(), key=lambda kv: kv[1]))
del sorted_tp_dict["Grades"]
ind = np.arange(len(sorted_tp_dict))
plt.figure(figsize=(20,5))
p1 = plt.bar(ind, list(sorted_tp_dict.values()))
plt.ylabel('Projects')
plt.title('% of projects aproved Project_grade wise')
plt.xticks(ind, list(sorted_tp_dict.keys()))
plt.show()
In [101]:
# we use count vectorizer to convert the values into one hot encoded features
vectorizer = CountVectorizer(vocabulary=list(sorted_tp_dict.keys()), lowercase=False, binary=True)
vectorizer.fit(project_data['project_grade_category'].values)
print(vectorizer.get_feature_names())
sub_categories_one_hot_3 = vectorizer.transform(project_data['project_grade_category'].values)
print("Shape of matrix after one hot encodig ",sub_categories_one_hot_3.shape)
1.4.2 Vectorizing Text data¶
1.4.2.1 Bag of words¶
In [102]:
# We are considering only the words which appeared in at least 10 documents(rows or projects).
vectorizer = CountVectorizer(min_df=10)
text_bow = vectorizer.fit_transform(preprocessed_essays)
print("Shape of matrix after one hot encodig ",text_bow.shape)
1.4.2.2 Bag of Words on `project_title`
In [103]:
# you can vectorize the title also
# before you vectorize the title make sure you preprocess it
vectorizer1 = CountVectorizer()
text_bow1 = vectorizer1.fit_transform(preprocessed_titles)
print("Shape of matrix after one hot encodig ",text_bow1.shape)
In [104]:
# Similarly you can vectorize for title also
1.4.2.3 TFIDF vectorizer¶
In [105]:
from sklearn.feature_extraction.text import TfidfVectorizer
vectorizer = TfidfVectorizer(min_df=10)
text_tfidf = vectorizer.fit_transform(preprocessed_essays)
print("Shape of matrix after one hot encodig ",text_tfidf.shape)
1.4.2.4 TFIDF Vectorizer on `project_title`
In [106]:
# Similarly you can vectorize for title also
from sklearn.feature_extraction.text import TfidfVectorizer
vectorizer2 = TfidfVectorizer()
text_tfidf2 = vectorizer2.fit_transform(preprocessed_titles)
print("Shape of matrix after one hot encodig ",text_tfidf2.shape)
1.4.2.5 Using Pretrained Models: Avg W2V¶
In [107]:
'''
# Reading glove vectors in python: https://stackoverflow.com/a/38230349/4084039
def loadGloveModel(gloveFile):
print ("Loading Glove Model")
f = open(gloveFile,'r', encoding="utf8")
model = {}
for line in tqdm(f):
splitLine = line.split()
word = splitLine[0]
embedding = np.array([float(val) for val in splitLine[1:]])
model[word] = embedding
print ("Done.",len(model)," words loaded!")
return model
model = loadGloveModel('glove.42B.300d.txt')
# ============================
Output:
Loading Glove Model
1917495it [06:32, 4879.69it/s]
Done. 1917495 words loaded!
# ============================
words = []
for i in preproced_texts:
words.extend(i.split(' '))
for i in preproced_titles:
words.extend(i.split(' '))
print("all the words in the coupus", len(words))
words = set(words)
print("the unique words in the coupus", len(words))
inter_words = set(model.keys()).intersection(words)
print("The number of words that are present in both glove vectors and our coupus", \
len(inter_words),"(",np.round(len(inter_words)/len(words)*100,3),"%)")
words_courpus = {}
words_glove = set(model.keys())
for i in words:
if i in words_glove:
words_courpus[i] = model[i]
print("word 2 vec length", len(words_courpus))
# stronging variables into pickle files python: http://www.jessicayung.com/how-to-use-pickle-to-save-and-load-variables-in-python/
import pickle
with open('glove_vectors', 'wb') as f:
pickle.dump(words_courpus, f)
'''
Out[107]:
In [110]:
# stronging variables into pickle files python: http://www.jessicayung.com/how-to-use-pickle-to-save-and-load-variables-in-python/
# make sure you have the glove_vectors file
with open('glove_vectors', 'rb') as f:
model = pickle.load(f)
glove_words = set(model.keys())
In [111]:
# average Word2Vec
# compute average word2vec for each review.
avg_w2v_vectors = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(preprocessed_essays): # for each review/sentence
vector = np.zeros(300) # as word vectors are of zero length
cnt_words =0; # num of words with a valid vector in the sentence/review
for word in sentence.split(): # for each word in a review/sentence
if word in glove_words:
vector += model[word]
cnt_words += 1
if cnt_words != 0:
vector /= cnt_words
avg_w2v_vectors.append(vector)
print(len(avg_w2v_vectors))
print(len(avg_w2v_vectors[0]))
1.4.2.6 Using Pretrained Models: AVG W2V on `project_title`
In [112]:
# Similarly you can vectorize for title also
avg_w2v_vectors1 = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(preprocessed_titles): # for each review/sentence
vector = np.zeros(300) # as word vectors are of zero length
cnt_words =0; # num of words with a valid vector in the sentence/review
for word in sentence.split(): # for each word in a review/sentence
if word in glove_words:
vector += model[word]
cnt_words += 1
if cnt_words != 0:
vector /= cnt_words
avg_w2v_vectors1.append(vector)
print(len(avg_w2v_vectors1))
print(len(avg_w2v_vectors1[0]))
1.4.2.7 Using Pretrained Models: TFIDF weighted W2V¶
In [113]:
# S = ["abc def pqr", "def def def abc", "pqr pqr def"]
tfidf_model = TfidfVectorizer()
tfidf_model.fit(preprocessed_essays)
# we are converting a dictionary with word as a key, and the idf as a value
dictionary = dict(zip(tfidf_model.get_feature_names(), list(tfidf_model.idf_)))
tfidf_words = set(tfidf_model.get_feature_names())
In [114]:
# average Word2Vec
# compute average word2vec for each review.
tfidf_w2v_vectors = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(preprocessed_essays): # for each review/sentence
vector = np.zeros(300) # as word vectors are of zero length
tf_idf_weight =0; # num of words with a valid vector in the sentence/review
for word in sentence.split(): # for each word in a review/sentence
if (word in glove_words) and (word in tfidf_words):
vec = model[word] # getting the vector for each word
# here we are multiplying idf value(dictionary[word]) and the tf value((sentence.count(word)/len(sentence.split())))
tf_idf = dictionary[word]*(sentence.count(word)/len(sentence.split())) # getting the tfidf value for each word
vector += (vec * tf_idf) # calculating tfidf weighted w2v
tf_idf_weight += tf_idf
if tf_idf_weight != 0:
vector /= tf_idf_weight
tfidf_w2v_vectors.append(vector)
print(len(tfidf_w2v_vectors))
print(len(tfidf_w2v_vectors[0]))
1.4.2.9 Using Pretrained Models: TFIDF weighted W2V on `project_title`
In [115]:
# Similarly you can vectorize for title also
In [116]:
# S = ["abc def pqr", "def def def abc", "pqr pqr def"]
tfidf_model = TfidfVectorizer()
tfidf_model.fit(preprocessed_titles)
# we are converting a dictionary with word as a key, and the idf as a value
dictionary = dict(zip(tfidf_model.get_feature_names(), list(tfidf_model.idf_)))
tfidf_words = set(tfidf_model.get_feature_names())
In [117]:
# average Word2Vec
# compute average word2vec for each review.
tfidf_w2v_vectors1 = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(preprocessed_titles): # for each review/sentence
vector = np.zeros(300) # as word vectors are of zero length
tf_idf_weight =0; # num of words with a valid vector in the sentence/review
for word in sentence.split(): # for each word in a review/sentence
if (word in glove_words) and (word in tfidf_words):
vec = model[word] # getting the vector for each word
# here we are multiplying idf value(dictionary[word]) and the tf value((sentence.count(word)/len(sentence.split())))
tf_idf = dictionary[word]*(sentence.count(word)/len(sentence.split())) # getting the tfidf value for each word
vector += (vec * tf_idf) # calculating tfidf weighted w2v
tf_idf_weight += tf_idf
if tf_idf_weight != 0:
vector /= tf_idf_weight
tfidf_w2v_vectors1.append(vector)
print(len(tfidf_w2v_vectors1))
print(len(tfidf_w2v_vectors1[0]))
1.4.3 Vectorizing Numerical features¶
In [118]:
# check this one: https://www.youtube.com/watch?v=0HOqOcln3Z4&t=530s
# standardization sklearn: https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.StandardScaler.html
from sklearn.preprocessing import StandardScaler
# price_standardized = standardScalar.fit(project_data['price'].values)
# this will rise the error
# ValueError: Expected 2D array, got 1D array instead: array=[725.05 213.03 329. ... 399. 287.73 5.5 ].
# Reshape your data either using array.reshape(-1, 1)
price_scalar = StandardScaler()
price_scalar.fit(project_data['price'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
print(f"Mean : {price_scalar.mean_[0]}, Standard deviation : {np.sqrt(price_scalar.var_[0])}")
# Now standardize the data with above maen and variance.
price_standardized = price_scalar.transform(project_data['price'].values.reshape(-1, 1))
In [119]:
price_standardized
Out[119]:
In [120]:
teacher_scalar = StandardScaler()
teacher_scalar.fit(project_data['teacher_number_of_previously_posted_projects'].values.reshape(-1,1))
print(f"Mean : {teacher_scalar.mean_[0]}, Standard deviation : {np.sqrt(teacher_scalar.var_[0])}")
In [121]:
# Now standardize the data with above maen and variance.
teacher_standardized = teacher_scalar.transform(project_data['teacher_number_of_previously_posted_projects'].values.reshape(-1, 1))
In [122]:
teacher_standardized
Out[122]:
1.4.4 Merging all the above features¶
- we need to merge all the numerical vectors i.e catogorical, text, numerical vectors
In [123]:
print(categories_one_hot.shape)
print(sub_categories_one_hot.shape)
print(text_bow.shape)
print(price_standardized.shape)
In [124]:
# merge two sparse matrices: https://stackoverflow.com/a/19710648/4084039
from scipy.sparse import hstack
# with the same hstack function we are concatinating a sparse matrix and a dense matirx :)
X = hstack((categories_one_hot, sub_categories_one_hot, text_bow, price_standardized))
X.shape
Out[124]:
Assignment 2: Apply TSNE
If you are using any code snippet from the internet, you have to provide the reference/citations, as we did in the above cells. Otherwise, it will be treated as plagiarism without citations.
- In the above cells we have plotted and analyzed many features. Please observe the plots and write the observations in markdown cells below every plot.
- EDA: Please complete the analysis of the feature: teacher_number_of_previously_posted_projects
-
- Build the data matrix using these features
- school_state : categorical data (one hot encoding)
- clean_categories : categorical data (one hot encoding)
- clean_subcategories : categorical data (one hot encoding)
- teacher_prefix : categorical data (one hot encoding)
- project_grade_category : categorical data (one hot encoding)
- project_title : text data (BOW, TFIDF, AVG W2V, TFIDF W2V)
- price : numerical
- teacher_number_of_previously_posted_projects : numerical
- Now, plot FOUR t-SNE plots with each of these feature sets.
- categorical, numerical features + project_title(BOW)
- categorical, numerical features + project_title(TFIDF)
- categorical, numerical features + project_title(AVG W2V)
- categorical, numerical features + project_title(TFIDF W2V)
- Concatenate all the features and Apply TNSE on the final data matrix
- Note 1: The TSNE accepts only dense matrices
- Note 2: Consider only 5k to 6k data points to avoid memory issues. If you run into memory error issues, reduce the number of data points but clearly state the number of datat-poins you are using
In [125]:
# this is the example code for TSNE
import numpy as np
from sklearn.manifold import TSNE
from sklearn import datasets
import pandas as pd
import matplotlib.pyplot as plt
iris = datasets.load_iris()
x = iris['data']
y = iris['target']
tsne = TSNE(n_components=2, perplexity=30, learning_rate=200)
X_embedding = tsne.fit_transform(x)
# if x is a sparse matrix you need to pass it as X_embedding = tsne.fit_transform(x.toarray()) , .toarray() will convert the sparse matrix into dense matrix
for_tsne = np.hstack((X_embedding, y.reshape(-1,1)))
for_tsne_df = pd.DataFrame(data=for_tsne, columns=['Dimension_x','Dimension_y','Score'])
colors = {0:'red', 1:'blue', 2:'green'}
plt.scatter(for_tsne_df['Dimension_x'], for_tsne_df['Dimension_y'], c=for_tsne_df['Score'].apply(lambda x: colors[x]))
plt.show()
In [155]:
#hstack with all features including titles text_bow(w2v),text_tfidf,avg_w2v, avg_tfidf
m = hstack((sub_categories_one_hot_1[0:4000],categories_one_hot[0:4000],sub_categories_one_hot[0:4000],sub_categories_one_hot_2[0:4000],sub_categories_one_hot_3[0:4000],text_bow1[0:4000],text_tfidf2[0:4000],avg_w2v_vectors1[0:4000],tfidf_w2v_vectors1[0:4000],teacher_standardized[0:4000]))
m.shape
Out[155]:
In [134]:
m.dtype
Out[134]:
In [156]:
d = hstack((sub_categories_one_hot_1[0:4000],categories_one_hot[0:4000],sub_categories_one_hot[0:4000],sub_categories_one_hot_2[0:4000],sub_categories_one_hot_3[0:4000],text_bow1[0:4000],teacher_standardized[0:4000]))
d.shape
Out[156]:
In [157]:
d1=hstack((sub_categories_one_hot_1[0:4000],categories_one_hot[0:4000],sub_categories_one_hot[0:4000],sub_categories_one_hot_2[0:4000],sub_categories_one_hot_3[0:4000],text_tfidf2[0:4000],teacher_standardized[0:4000]))
d1.shape
Out[157]:
In [159]:
d2=hstack((sub_categories_one_hot_1[0:4000],categories_one_hot[0:4000],sub_categories_one_hot[0:4000],sub_categories_one_hot_2[0:4000],sub_categories_one_hot_3[0:4000],avg_w2v_vectors1[0:4000],teacher_standardized[0:4000]))
d2.shape
Out[159]:
In [160]:
d3=hstack((sub_categories_one_hot_1[0:4000],categories_one_hot[0:4000],sub_categories_one_hot[0:4000],sub_categories_one_hot_2[0:4000],sub_categories_one_hot_3[0:4000],tfidf_w2v_vectors1[0:4000],teacher_standardized[0:4000]))
d3.shape
Out[160]:
2.1 TSNE with `BOW` encoding of `project_title` feature
In [170]:
# please write all of the code with proper documentation and proper titles for each subsection
# when you plot any graph make sure you use
# a. Title, that describes your plot, this will be very helpful to the reader
# b. Legends if needed
# c. X-axis label
# d. Y-axis label
import numpy as np
from sklearn.manifold import TSNE
from sklearn import datasets
import pandas as pd
import matplotlib.pyplot as plt
#iris = datasets.load_iris()
#x = d
x=text_bow1[0:4000]
y = project_data['project_is_approved'][0:4000]
tsne = TSNE(n_components=2, perplexity=30, learning_rate=500)
X_embedding = tsne.fit_transform(x.toarray())
# if x is a sparse matrix you need to pass it as X_embedding = tsne.fit_transform(x.toarray()) , .toarray() will convert the sparse matrix into dense matrix
plt.title('tSNE Plot for title BOW Vs Project approved state')
plt.xlabel('Multiple stack of BOW features')
plt.ylabel('Project is Approved')
for_tsne = np.hstack((X_embedding, y.as_matrix().reshape(-1,1)))
for_tsne_df = pd.DataFrame(data=for_tsne, columns=['Dimension_x','Dimension_y','Score'])
colors = {0:'red', 1:'blue', 2:'green'}
plt.scatter(for_tsne_df['Dimension_x'], for_tsne_df['Dimension_y'], c=for_tsne_df['Score'].apply(lambda x: colors[x]))
plt.show()
2.2 TSNE with `TFIDF` encoding of `project_title` feature
In [171]:
# please write all the code with proper documentation, and proper titles for each subsection
# when you plot any graph make sure you use
# a. Title, that describes your plot, this will be very helpful to the reader
# b. Legends if needed
# c. X-axis label
# d. Y-axis label
import numpy as np
from sklearn.manifold import TSNE
from sklearn import datasets
import pandas as pd
import matplotlib.pyplot as plt
#iris = datasets.load_iris()
#x = d1
x= text_tfidf2[0:4000]
y = project_data['project_is_approved'][0:4000]
tsne = TSNE(n_components=2, perplexity=30, learning_rate=500)
X_embedding = tsne.fit_transform(x.toarray())
# if x is a sparse matrix you need to pass it as X_embedding = tsne.fit_transform(x.toarray()) , .toarray() will convert the sparse matrix into dense matrix
plt.title('tSNE Plot for title TF-IDF Vs Project approved state')
plt.xlabel('Multiple stack of TF-IDF features')
plt.ylabel('Project is Approved')
for_tsne = np.hstack((X_embedding, y.as_matrix().reshape(-1,1)))
for_tsne_df = pd.DataFrame(data=for_tsne, columns=['Dimension_x','Dimension_y','Score'])
colors = {0:'red', 1:'blue', 2:'green'}
plt.scatter(for_tsne_df['Dimension_x'], for_tsne_df['Dimension_y'], c=for_tsne_df['Score'].apply(lambda x: colors[x]))
plt.show()
2.3 TSNE with `AVG W2V` encoding of `project_title` feature
In [172]:
# please write all the code with proper documentation, and proper titles for each subsection
# when you plot any graph make sure you use
# a. Title, that describes your plot, this will be very helpful to the reader
# b. Legends if needed
# c. X-axis label
# d. Y-axis label
import numpy as np
from sklearn.manifold import TSNE
from sklearn import datasets
import pandas as pd
import matplotlib.pyplot as plt
#iris = datasets.load_iris()
x=avg_w2v_vectors1[0:4000]
#x = d2
y = project_data['project_is_approved'][0:4000]
tsne = TSNE(n_components=2, perplexity=30, learning_rate=500)
X_embedding = tsne.fit_transform(np.array(x))
# if x is a sparse matrix you need to pass it as X_embedding = tsne.fit_transform(x.toarray()) , .toarray() will convert the sparse matrix into dense matrix
plt.title('tSNE Plot for title AVG w2v Vs Project approved state')
plt.xlabel('Multiple stack of AVG w2v features')
plt.ylabel('Project is Approved')
for_tsne = np.hstack((X_embedding, y.as_matrix().reshape(-1,1)))
for_tsne_df = pd.DataFrame(data=for_tsne, columns=['Dimension_x','Dimension_y','Score'])
colors = {0:'red', 1:'blue', 2:'green'}
plt.scatter(for_tsne_df['Dimension_x'], for_tsne_df['Dimension_y'], c=for_tsne_df['Score'].apply(lambda x: colors[x]))
plt.show()
2.4 TSNE with `TFIDF Weighted W2V` encoding of `project_title` feature
In [173]:
# please write all the code with proper documentation, and proper titles for each subsection
# when you plot any graph make sure you use
# a. Title, that describes your plot, this will be very helpful to the reader
# b. Legends if needed
# c. X-axis label
# d. Y-axis labelimport numpy as np
from sklearn.manifold import TSNE
from sklearn import datasets
import pandas as pd
import matplotlib.pyplot as plt
#iris = datasets.load_iris()
#x = d3
x=tfidf_w2v_vectors1[0:4000]
y = project_data['project_is_approved'][0:4000]
tsne = TSNE(n_components=2, perplexity=30, learning_rate=500)
plt.title('tSNE Plot for title TFIDF weighted w2v Vs Project approved state')
plt.xlabel('Multiple stack of TFIDF weighted w2v features')
plt.ylabel('Project is Approved')
X_embedding = tsne.fit_transform(np.array(x))
# if x is a sparse matrix you need to pass it as X_embedding = tsne.fit_transform(x.toarray()) , .toarray() will convert the sparse matrix into dense matrix
for_tsne = np.hstack((X_embedding, y.as_matrix().reshape(-1,1)))
for_tsne_df = pd.DataFrame(data=for_tsne, columns=['Dimension_x','Dimension_y','Score'])
colors = {0:'red', 1:'blue', 2:'green'}
plt.scatter(for_tsne_df['Dimension_x'], for_tsne_df['Dimension_y'], c=for_tsne_df['Score'].apply(lambda x: colors[x]))
plt.show()
TSNE with all features¶
In [174]:
# please write all of the code with proper documentation and proper titles for each subsection
# when you plot any graph make sure you use
# a. Title, that describes your plot, this will be very helpful to the reader
# b. Legends if needed
# c. X-axis label
# d. Y-axis label
import numpy as np
from sklearn.manifold import TSNE
from sklearn import datasets
import pandas as pd
import matplotlib.pyplot as plt
#iris = datasets.load_iris()
x = d
y = project_data['project_is_approved'][0:4000]
tsne = TSNE(n_components=2, perplexity=30, learning_rate=500)
X_embedding = tsne.fit_transform(x.toarray())
# if x is a sparse matrix you need to pass it as X_embedding = tsne.fit_transform(x.toarray()) , .toarray() will convert the sparse matrix into dense matrix
plt.title('tSNE Plot for stack of all features + title BOW Vs Project approved state')
plt.xlabel('Multiple stack of BOW features')
plt.ylabel('Project is Approved')
for_tsne = np.hstack((X_embedding, y.as_matrix().reshape(-1,1)))
for_tsne_df = pd.DataFrame(data=for_tsne, columns=['Dimension_x','Dimension_y','Score'])
colors = {0:'red', 1:'blue', 2:'green'}
plt.scatter(for_tsne_df['Dimension_x'], for_tsne_df['Dimension_y'], c=for_tsne_df['Score'].apply(lambda x: colors[x]))
plt.show()
In [175]:
# please write all the code with proper documentation, and proper titles for each subsection
# when you plot any graph make sure you use
# a. Title, that describes your plot, this will be very helpful to the reader
# b. Legends if needed
# c. X-axis label
# d. Y-axis label
import numpy as np
from sklearn.manifold import TSNE
from sklearn import datasets
import pandas as pd
import matplotlib.pyplot as plt
#iris = datasets.load_iris()
x = d1
y = project_data['project_is_approved'][0:4000]
tsne = TSNE(n_components=2, perplexity=30, learning_rate=500)
X_embedding = tsne.fit_transform(x.toarray())
# if x is a sparse matrix you need to pass it as X_embedding = tsne.fit_transform(x.toarray()) , .toarray() will convert the sparse matrix into dense matrix
plt.title('tSNE Plot for stack of all features + title TF-IDF Vs Project approved state')
plt.xlabel('Multiple stack of TF-IDF features')
plt.ylabel('Project is Approved')
for_tsne = np.hstack((X_embedding, y.as_matrix().reshape(-1,1)))
for_tsne_df = pd.DataFrame(data=for_tsne, columns=['Dimension_x','Dimension_y','Score'])
colors = {0:'red', 1:'blue', 2:'green'}
plt.scatter(for_tsne_df['Dimension_x'], for_tsne_df['Dimension_y'], c=for_tsne_df['Score'].apply(lambda x: colors[x]))
plt.show()
In [176]:
# please write all the code with proper documentation, and proper titles for each subsection
# when you plot any graph make sure you use
# a. Title, that describes your plot, this will be very helpful to the reader
# b. Legends if needed
# c. X-axis label
# d. Y-axis label
import numpy as np
from sklearn.manifold import TSNE
from sklearn import datasets
import pandas as pd
import matplotlib.pyplot as plt
#iris = datasets.load_iris()
x = d2
y = project_data['project_is_approved'][0:4000]
tsne = TSNE(n_components=2, perplexity=30, learning_rate=500)
X_embedding = tsne.fit_transform(x.toarray())
# if x is a sparse matrix you need to pass it as X_embedding = tsne.fit_transform(x.toarray()) , .toarray() will convert the sparse matrix into dense matrix
plt.title('tSNE Plot for stack of all features + title AVG w2v Vs Project approved state')
plt.xlabel('Multiple stack of AVG w2v features')
plt.ylabel('Project is Approved')
for_tsne = np.hstack((X_embedding, y.as_matrix().reshape(-1,1)))
for_tsne_df = pd.DataFrame(data=for_tsne, columns=['Dimension_x','Dimension_y','Score'])
colors = {0:'red', 1:'blue', 2:'green'}
plt.scatter(for_tsne_df['Dimension_x'], for_tsne_df['Dimension_y'], c=for_tsne_df['Score'].apply(lambda x: colors[x]))
plt.show()
In [177]:
# when you plot any graph make sure you use
# a. Title, that describes your plot, this will be very helpful to the reader
# b. Legends if needed
# c. X-axis label
# d. Y-axis labelimport numpy as np
from sklearn.manifold import TSNE
from sklearn import datasets
import pandas as pd
import matplotlib.pyplot as plt
#iris = datasets.load_iris()
x = d3
y = project_data['project_is_approved'][0:4000]
tsne = TSNE(n_components=2, perplexity=30, learning_rate=500)
plt.title('tSNE Plot for stack of all features + title TFIDF weighted w2v Vs Project approved state')
plt.xlabel('Multiple stack of TFIDF weighted w2v features')
plt.ylabel('Project is Approved')
X_embedding = tsne.fit_transform(x.toarray())
# if x is a sparse matrix you need to pass it as X_embedding = tsne.fit_transform(x.toarray()) , .toarray() will convert the sparse matrix into dense matrix
for_tsne = np.hstack((X_embedding, y.as_matrix().reshape(-1,1)))
for_tsne_df = pd.DataFrame(data=for_tsne, columns=['Dimension_x','Dimension_y','Score'])
colors = {0:'red', 1:'blue', 2:'green'}
plt.scatter(for_tsne_df['Dimension_x'], for_tsne_df['Dimension_y'], c=for_tsne_df['Score'].apply(lambda x: colors[x]))
plt.show()
In [178]:
# please write all the code with proper documentation, and proper titles for each subsection
# when you plot any graph make sure you use
# a. Title, that describes your plot, this will be very helpful to the reader
# b. Legends if needed
# c. X-axis label
# d. Y-axis label
import numpy as np
from sklearn.manifold import TSNE
from sklearn import datasets
import pandas as pd
import matplotlib.pyplot as plt
#iris = datasets.load_iris()
x = m
y = project_data['project_is_approved'][0:4000]
tsne = TSNE(n_components=2, perplexity=30, learning_rate=500)
X_embedding = tsne.fit_transform(x.toarray())
# if x is a sparse matrix you need to pass it as X_embedding = tsne.fit_transform(x.toarray()) , .toarray() will convert the sparse matrix into dense matrix
plt.title('tSNE Plot for stack of all features Vs Project approved state')
plt.xlabel('Multiple stack of all features')
plt.ylabel('Project is Approved')
for_tsne = np.hstack((X_embedding, y.as_matrix().reshape(-1,1)))
for_tsne_df = pd.DataFrame(data=for_tsne, columns=['Dimension_x','Dimension_y','Score'])
colors = {0:'red', 1:'blue', 2:'green'}
plt.scatter(for_tsne_df['Dimension_x'], for_tsne_df['Dimension_y'], c=for_tsne_df['Score'].apply(lambda x: colors[x]))
plt.show()
2.5 Summary
In [152]:
# Write few sentences about the results that you obtained and the observations you made.
Basic points on TSNE:¶
As we know that TSNE is used for dimensionality reduction and to preserve the neighbourhood of the data.
PCA is old and basic type of dimensionality reduction technique.
We opt TSNE over PCA because, PCA has a very huge loss of data when data is reduced from high dimensions.
There is also a major limitation in TSNE known as Crowding problem.
Summary on results obtained:¶
From the above plotting we can understand that based on the perplexity and learning rate we can obtain a good shape to the data.
Even considering the 4000 sample data points BOW and TFIDF including all features has atmost managed to get a proper shape at the end of the compilation when compared to the avg BOW and avg TFIDF.
In [ ]: